A matching algorithm for motion analysis of dense populations

Dmitry Chetverikov, Attila Lerch

Research output: Contribution to journalArticle

5 Citations (Scopus)


This study deals with the estimation of the motion statistics of spermatozoa populations. We consider this application task as a test problem in a more general framework which is the matching of two feature point sets for motion analysis of a dense ensemble of independent, intensively moving small-size objects. The direction of motion of each object can be predicted from the object shape, while the velocity magnitude is unknown. Motion across the edges of the field is significant Due to low visibility, the objects can appear and disappear in any part of the viewfield. A fast and effective matching algorithm is proposed which is based on the hypothesis testing principle and uses the directional coherence heuristic. Results of some experiments with the proposed algorithm are shown and discussed.

Original languageEnglish
Pages (from-to)743-749
Number of pages7
JournalPattern Recognition Letters
Issue number11
Publication statusPublished - Nov 1990


  • Correspondence
  • directional coherence
  • hypothesis testing
  • motion object matching

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'A matching algorithm for motion analysis of dense populations'. Together they form a unique fingerprint.

  • Cite this